Health plan dissatisfaction was higher among Medicaid managed long-term services and supports (MLTSS) beneficiaries who did not follow through with an intention to change health plans.
Objectives: This paper examines (1) the rate of plan switching among beneficiaries enrolled in a Medicaid managed long-term services and supports (MLTSS) program in Virginia, (2) barriers that prevent beneficiaries from changing plans, and (3) the extent to which a change in plans is associated with greater satisfaction with the current health plan.
Study Design: Survey data from a representative sample of 1048 members enrolled in Commonwealth Coordinated Care Plus, a Virginia Medicaid MLTSS program.
Methods: The survey ascertained whether beneficiaries changed plans at the previous open enrollment period, whether they wanted to change plans but did not, and reasons for not following through with a plan change. Logistic regression analysis examined the association between the intention to change plans and satisfaction with the current health plan.
Results: Seven percent of respondents changed plans during the previous open enrollment. However, twice as many respondents (15%) wanted to change plans but did not. The main reason for not changing plans was uncertainty about whether the new plan would meet their needs better than their current plan. Logistic regression analysis shows that an intention to change plans (realized or not) was associated with higher odds (3.5 times higher) of being dissatisfied with the current health plan compared with beneficiaries who had no intention to change plans.
Conclusions: Greater dissatisfaction after a recent plan change may indicate that these members have specific needs beyond the scope of services offered by managed care organizations.
Am J Manag Care. 2022;28(12):e428-e435. https://doi.org/10.37765/ajmc.2022.89278
Increasing consumer choice has been a central feature of many Medicaid managed care programs.1 Although there may be little or no difference in cost-sharing among Medicaid managed care plans, beneficiaries may choose plans based on several other factors, such as differences in the size, composition, and accessibility of provider networks; customer service; quality of care coordination services; and optional services.2 By allowing beneficiaries to choose from several competing managed care organizations (MCOs), the expectation is that MCOs will be incentivized to provide high-quality services to members.3
Medicaid beneficiaries in managed long-term services and supports (MLTSS) programs may find it especially difficult to navigate the MCO choices available to them, given the complexity of health and personal care services that they require to continue living in the community. MLTSS programs may include not only those dually eligible for Medicare and Medicaid, but also often other beneficiaries with complex health needs, including those with physical and intellectual disabilities.4,5 Beneficiaries and their caregivers must decide on the MCO that is best suited to provide the support services, in addition to health services, that will enable them to continue living in the community.
Research has indicated that greater choice of plans among Medicaid beneficiaries can lead to improved satisfaction with care and greater perceived access to services, in part because of the ability of beneficiaries to change to plans that offer higher-quality services or are better suited to their needs.6,7 Therefore, some degree of plan switching is both expected and desirable. Although a high rate of plan switching may disrupt continuity of care among beneficiaries, low rates of plan switching may indicate either a lack of clear choices or difficulty in making choices among complex alternatives.8,9 If the latter is true, then the actual rate of plan switching may underestimate the extent to which MLTSS members desire a plan change but are unable or unwilling to follow through with a change.
Most prior research on plan switching has focused on rates of plan switching among beneficiaries of Medicare Advantage, commercially insured populations, and non-MLTSS Medicaid managed care populations, but there has been little or no research on plan switching among Medicaid MLTSS beneficiaries to the best of our knowledge. Not surprisingly, the desire for lower premiums is the main factor behind plan switching in commercial insurance.10 About 10% of Medicare Advantage enrollees switch plans annually—a much lower rate compared with those enrolled in Affordable Care Act marketplace plans—although plan switching tended to be higher among younger enrollees and dual-eligible enrollees.11 Few studies have assessed the role of health status in plan switching, an important consideration for MLTSS populations. Among Medicare Advantage enrollees, those enrolled in Special Needs Plans (who include many with disabilities and complex health needs) were less likely to switch plans compared with other dual-eligible enrollees.11 In a study of individuals enrolled in employer-sponsored insurance coverage, those with chronic conditions who were dissatisfied with their health plans were less likely to change plans compared with those who did not have chronic conditions.12 The latter study is especially relevant for MLTSS populations, as it suggests that dissatisfaction with the health plan alone may be an insufficient motivating factor for those with significant health needs, as well as the possibility of greater barriers to plan switching when the current plan is not meeting their needs.
The objective of this paper is to examine plan switching among Medicaid beneficiaries enrolled in Commonwealth Coordinated Care (CCC) Plus, a Medicaid MLTSS program in Virginia that began in 2017 and serves approximately 244,000 beneficiaries.13 A central feature of CCC Plus is that long-term services and supports are provided through 1 of 6 MCOs for members requiring such services.14 MCOs must provide a standardized set of health, personal care, and long-term care services with no cost sharing, but they may differ based on their network of providers, quality-of-care metrics, care coordination services, customer service experiences, and expanded benefits. Beneficiaries who do not initially choose an MCO plan are auto-enrolled and then have 90 days after benefits begin to request a change in plans. Also, beneficiaries can change plans during an open enrollment period during October through December of each year.
This research addresses a number of gaps in the research literature on plan switching, primarily by being the first study to our knowledge to examine plan switching among an MLTSS population. In addition, the study examines both an intention to change plans and completed plan changes, and it examines the reasons when intended plan changes are not completed. Specifically, the research examines (1) the extent to which CCC Plus members wanted to change health plans in the previous year; (2) the extent to which beneficiaries did not change plans when they wanted to, and the reasons for not following through on an intention to change plans; and (3) the extent to which a desire to change plans—whether or not such changes actually occurred—is associated with greater satisfaction with the health plan they were enrolled in at the time of the survey.
CCC Plus Member Survey
A survey of Virginia Medicaid CCC Plus members was conducted between August and November 2019. The primary objectives of the survey included assessing members’ experiences with care coordinators, health plans, and barriers to using health, personal care, and social services. The survey also included questions on changes or intended changes in health plans in the past 12 months, reasons for these changes, and intentions to change health plans in the future. Other information was obtained on CCC Plus members’ unmet needs; social needs, such as food and housing insecurity; and other self-ratings of physical and mental health, general well-being, and difficulties with activities of daily living.
The survey utilized a representative sample of noninstitutionalized members who had 6 months or more of experience with the CCC Plus program. The sample frame excluded members residing in nursing facilities, members who were deceased, and members who did not speak English as their primary language (less than 1% of members). It was fielded to 3600 members who were randomly selected from CCC Plus enrollment files and yielded a 30% response rate, for a final sample of 1048 respondents. All respondents voluntarily participated in this study. An analysis of differences between survey respondents and nonrespondents showed some differences by age but little difference by gender, race/ethnicity, or region. A survey weight was constructed that corrects for differential nonresponse based on these factors and was used for all analyses in this paper. More than 40% of surveys were completed by proxy (eg, relative, guardian, friend, personal care attendant) if the member was not able to complete the survey. The analysis accounts for potential differences in reporting between sampled members and proxies (as discussed). All measures included in the analysis are derived from the survey data, with the exception of the measure of duration of enrollment in the current health plan, which was extracted from enrollment data. A copy of the survey questionnaire is included as the eAppendix (available at ajmc.com).
The institutional review board at Virginia Commonwealth University reviewed and approved this study as required by the contract with the Virginia Department of Medical Assistance Services to evaluate Medicaid plans.
Health Plan Switching
Survey respondents were asked: “You had an opportunity to switch health plans during the last open enrollment period. Did you switch health plans?” The survey included 4 responses: (1) “No, I did not switch to a new health plan during the last open enrollment,” (2) “Yes, I switched to a new health plan during the last open enrollment,” (3) “I wanted to switch to a new health plan but did not switch,” and (4) “I don’t remember.” There were 124 (12%) respondents who reported that they did not recall whether they switched plans or responded with no or multiple choices. We categorized the responses into (1) changed plan at the open enrollment and (2) did not change or had no intention to change their plan at the open enrollment and unknown responses. For those who reported that they wanted to but did not switch health plans, a follow-up question was asked to determine the reasons for not following through with a switch.
Satisfaction With Health Plan
The analysis includes 2 measures of satisfaction with current health plan, including a measure of overall satisfaction with health plan and the likelihood of changing health plans during the next open enrollment period. For the overall measure of plan satisfaction, participants were asked, “Overall, how satisfied are you with the services you are receiving through their health plan?” Respondents were given the options of “very satisfied,” “satisfied,” and “not satisfied.” Although only a single response indicated negative experience with the health plan, the percentage reporting “not satisfied” (9.3%) was similar to estimates based on a 4-point scale used in the Consumer Assessment of Healthcare Providers and Systems survey—that is, similar when combining responses from the 2 categories representing negative experiences.15
For this analysis, a binary measure was created where respondents were coded as 1 if they reported not being satisfied with their health plan and 0 for all other responses. A small number of missing responses (3.7%) for this question were omitted from the analysis.
A second measure of plan satisfaction assesses likelihood of changing their current plan. Beneficiaries were asked, “How likely is it that you will change your Medicaid health plan during the next open enrollment?” The options were “very likely,” “somewhat likely,” and “unlikely.” For this study, responses were coded as 1 if the beneficiaries responded “very likely” or “somewhat likely” and 0 if they reported otherwise. There were 7.2% of members who did not respond to this question and were omitted from the analysis for this measure.
We controlled for other member characteristics, such as age, education, gender, race/ethnicity, employment status, difficulty with activities of daily living, housing security, living arrangements, number of close relationships, number of chronic diseases, self-efficacy regarding their health, respondents’ role (ie, self vs proxy), and length of time in the health plan.
Self-efficacy refers to an individual’s belief in their capacity to execute behaviors necessary to produce specific performance attainments.16 Because of space constraints in the survey questionnaire, we did not use validated self-efficacy measures, such as the PROMIS measures of self-efficacy for managing chronic conditions.17 Instead, self-efficacy was measured based on 4 questions developed specifically for this population: (1) whether they feel confident that they can use medications correctly, (2) whether they reported that caring for their health was manageable, (3) whether they reported having all the support needed for daily care of themselves, and (4) whether they can tell when a health problem needs attention right away. Respondents had a choice of “strongly agree,” “agree,” “disagree,” and “strongly disagree.” We constructed an index of self-efficacy based on a sum of the responses to these 4 questions, with 12 indicating the highest level of self-efficacy. A Cronbach’s α of 0.90 for the measures in the index indicated high internal consistency and validity. Nevertheless, we acknowledge that the measure is more limited and may not capture all the dimensions of self-efficacy relative to other established measures.
The survey asked respondents to indicate their number of close relationships by choosing from 4 categories: “none,” ”1 or 2,” “3 to 5,” and “more than 5.” For the purpose of analysis, we recategorized respondents as having no or at least 1 close relationship with their family or friends.
Duration of enrollment in current health plan was derived from member enrollment data and is measured as a binary indicator with 1 indicating that they have been enrolled in their current health plan for 6 months or longer. The regression analysis also controls for potential differences in survey responses based on self-reporting vs proxy reporting by including an indicator for this as an independent variable. To account for differences among individual MCOs in beneficiary experiences, a categorical measure identifying enrollment in the 6 MCOs at the time of the survey was also included, with MCO identities masked (labeled as MCO A through F).
We performed a descriptive analysis of sample characteristics, stratified by those who switched or wanted to switch plans in the past year and those who did not want to switch plans. We examined the percentage who changed plans, the percentage who wanted to change but did not complete a plan change, and the reasons for not completing a plan change.
We conducted 2 logistic regression analyses. The first regression model examined the likelihood of a respondent being dissatisfied with the current health plan. The second model evaluated the likelihood of a respondent wanting to change health plans at open enrollment. Independent variables in both models include a categorical measure for plan switching in the past year, with categories of (1) no intention to change plans (the reference group), (2) changed plans in the past year, and (3) did not complete an intended plan change. All other covariates were included in the analysis, including indicators identifying enrollment in the 6 MCOs. For ease of interpretation, we included the analysis of the marginal effects of the probability of being dissatisfied with the current plan and the probability of interest in switching plans during the next open enrollment. We used Stata version 13 (StataCorp) to perform all the analyses.
Frequency of and Reasons for Switching Health Plans
Table 1 presents findings on plan changes during the prior year, interest in changing plans during the next open enrollment period, and satisfaction with the current plan. Only 6.6% of respondents reported actually changing plans, but a much larger number (22%) reported that they did not complete an intended plan change. Reasons for not completing the plan change are listed in Figure 1. The most frequently cited reason for not completing a plan change was concern that the new health plan would not be better than the current plan (more than 60%). Smaller percentages identified not knowing how to switch plans (18%), the difficulty or time it took to switch plans (13%), and not being aware of the choice to change plans (12%).
Most survey respondents were either very satisfied (39.8%) or satisfied (46.7%) with their current health plan. Similarly, only 8% of respondents reported that they were very likely to change their current health plan during the next open enrollment, whereas 73% were unlikely to change their current health plan. Among those who indicated that they want to switch plans at the next open enrollment, the most frequently cited reason was to get better coverage for dental and vision care (29%) (Figure 2), which are both enhanced benefits that may be offered by health plans but are not required covered services for all populations.
Differences Between Those Wanting and Those Not Wanting to Switch Plans
Notable differences existed between respondents who had an intention to change health plans (whether completed or not) during the previous open enrollment period and respondents who did not have an intention to change plans. Compared with respondents who did not intend to change plans, those who intended to change plans were more likely to be non-Hispanic Black (43.2%), have less housing security (15.7%), have fewer social relationships (10.2%), have 4 or more health conditions (22.1%), have lower self-efficacy (mean score, 8.5), and be enrolled in the MCO labeled as MCO A (17.3%) (Table 2 [part A and part B]).
Logistic Regression Analysis
The results of the logistic regressions show that the odds of being dissatisfied with the current health plan were 3.5 times (P < .01) higher in members who had previously changed their health plan compared with members who did not change their health plan in the previous enrollment period (Table 3 [part A and part B]). The odds of being dissatisfied with the current health plan were 4.2 times (P < .001) higher in members who did not complete a plan change compared with members who did not intend to change their health plan in the previous enrollment period (Table 3).
The results also showed that housing status and self-efficacy were associated with the likelihood of being dissatisfied with the current plan. The odds of being dissatisfied with the current health plan were 3.4 times (P < .01) higher in members who did not have housing compared with members who had housing. Members with less self-efficacy were more likely to be dissatisfied with their health plan (odds ratio [OR], 0.7; P < .001).
The odds of wanting to change plans at the next open enrollment were almost 5 times (P < .001) higher in members who previously changed their health plan compared with members who did not intend to change their health plan in the previous enrollment period. Also, the odds of wanting to change one’s health plan at the next open enrollment were 4.2 times (P < .001) higher in members who did not complete a plan change compared with those with no intention of changing plans (Table 3). Other factors associated with a higher likelihood of changing plans include lower self-efficacy (OR, 0.9; P < .001) and having been enrolled in their plan longer (OR, 1.7; P < .05) (Table 3).
This study examined plan switching behavior among beneficiaries enrolled in CCC Plus, a Medicaid MLTSS program in Virginia, and assessed the extent to which a prior change in plan is associated with greater satisfaction with their current health plan. Relatively few respondents reported changing plans in the previous year (6.6%). Although this may suggest stability and high satisfaction with plans among members, a much higher percentage of respondents (about 22%) did not complete an intention to change. Most cited lack of understanding of how the alternative plans would differ from their current plan and uncertainty about the process of switching plans as the reasons for not following through on their interest to change. Relative to the majority of respondents who had no desire to change plans, those who did not complete a plan change were much more likely to be dissatisfied with their current plan.
Somewhat surprisingly, those who actually switched plans in the previous year experienced the same levels of dissatisfaction with their current plan as those who did not complete an intended plan change. This may reflect a lack of differences between plans on aspects of quality of care that are important to them or that they are in need of services or providers that are generally not covered through the program. The most frequently cited reason for wanting to change plans at the next open enrollment was an interest in getting better dental or vision coverage (29%), which is an optional but not mandatory benefit that has only limited coverage by some MCOs.
That lower self-efficacy was associated with both a desire to change plans and lower satisfaction with the current plan is also instructive. Self-efficacy is an essential factor for making a decision, especially for individuals with complex health needs, who are more likely to need help with understanding health plan materials and navigating the complex array of services and benefits that they need for their health and personal care.18 Most consumers have a basic understanding of their health care options but much less understanding of the more detailed and complex issues involved.19 Lack of understanding of health plan benefits has been found to be among the most important factors in causing a change in plans.20 This becomes even more problematic when beneficiaries lack social support and confidence in their ability to make decisions.
This study builds on prior studies by examining plan switching behavior among members enrolled in an MLTSS program, who generally have poorer health and more complex care needs relative to those enrolled in commercial health insurance or general Medicare and Medicaid. Whereas most other studies focus on actual plan changes as the main indicator of consumer intentions, the results from this study show that the number who desire to change plans is much greater than the number who actually switch plans, in part because members do not perceive differences among the plans in ways that are relevant to their needs. As a result, rates of plan disenrollment among MLTSS members may not be strong indicators of health plan quality, as suggested by prior studies that have examined differences in plan switching by health status.12 Although previous studies have shown that premiums and other costs affect satisfaction and decisions to change plans among enrollees in commercial health plans,21 these decisions are driven more by a fulfillment of health needs and access to particular providers when cost considerations are largely absent.22 Furthermore, switching to a different health plan does not necessarily lead to greater satisfaction with the plan if the alternatives do not provide services needed by enrollees, such as dental and vision benefits not required by Virginia Medicaid at the time of the survey.
This study is cross-sectional and therefore causality between plan changes and plan satisfaction cannot be firmly established. In addition, all information is based on self-report by survey respondents, which is subject to response error. Recall may be an even greater problem for a sample consisting of a higher proportion of older adults and individuals with disabilities, although this is offset to some extent by the use of proxy respondents for more than 40% of completed surveys, based on the person most familiar with the beneficiary’s health and personal situation. Also, although our measure of self-efficacy was strongly associated with health plan satisfaction and the desire to change plans, our measure was more limited than more established measures of self-efficacy due to survey questionnaire constraints. Finally, this study is based on a single MLTSS program in Virginia and may not be generalizable to members of MLTSS programs in other states.
This study is the first study to our knowledge to examine plan changing behavior among an MLTSS population. In general, the results suggest a high degree of satisfaction with health plans and a low rate of actual plan switching. However, a larger minority of beneficiaries are dissatisfied with their plans and would like to change but are unable or unwilling to do so, in part because they do not perceive how the plans differ in a way that is meaningful for their needs. Those wanting to change plans appear to be especially vulnerable, including having low self-efficacy, lower social support, and greater health and personal care needs. It is possible that they have specific needs that are beyond the scope of the CCC Plus program, and therefore acting on the choices available to them will not improve their experience. However, these beneficiaries may require more targeted assistance in understanding the options and making decisions about which MCOs may be best suited for their needs.
Author Affiliations: Health Behavior and Policy Department, Virginia Commonwealth University (SS, HS, PC), Richmond, VA; Virginia Department of Medical Assistance Services (LW), Richmond, VA.
Source of Funding: Virginia Department of Medical Assistance Services (Interagency Agreement IAG-407, Commonwealth Coordinated Care Plus, Project #BRN70024).
Author Disclosures: Drs Salehian, Saunders, and Cunningham report receiving grants from the Virginia Department of Medical Assistance Services, which administers the program examined in this paper. Dr Walker reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SS, HS, LW, PC); acquisition of data (SS, HS, PC); analysis and interpretation of data (SS, HS, LW, PC); drafting of the manuscript (SS, HS, PC); critical revision of the manuscript for important intellectual content (HS, LW, PC); statistical analysis (SS, PC); provision of patients or study materials (SS, HS); obtaining funding (PC); administrative, technical, or logistic support (SS); and supervision (PC).
Address Correspondence to: Shiva Salehian, MD, PhD, MPH, Virginia Commonwealth University, 830 E Main St, Richmond, VA 23219. Email: firstname.lastname@example.org.
1. Enrollment process for Medicaid managed care. Medicaid and CHIP Payment and Access Commission. Accessed September 9, 2021. https://www.macpac.gov/subtopic/enrollment-process-for-medicaid-managed-care/
2. Managed care’s effect on outcomes. Medicaid and CHIP Payment and Access Commission. Accessed September 9, 2021. https://www.macpac.gov/subtopic/managed-cares-effect-on-outcomes/
3. Hinton E, Stolyar L. 10 things to know about Medicaid managed care. Kaiser Family Foundation. February 23, 2022. Accessed November 11, 2022. https://www.kff.org/medicaid/issue-brief/10-things-to-know-about-medicaid-managed-care/
4. Schur CL, Berk ML. Choice of health plan: implications for access and satisfaction. Health Care Financ Rev. 1998;20(1):29-43.
5. Managed care. Medicaid. Accessed September 9, 2021. https://www.medicaid.gov/medicaid/managed-care/index.html
6. Reese S. Disenrollment. what it costs, how to stop it. Bus Health. 1997;15(10):40-44.
7. Buchmueller TC, Gilmer T, Harris K. Health plan disenrollment in a choice-based Medicaid managed care program. Inquiry. 2004;41(4):447-460. doi:10.5034/inquiryjrnl_41.4.447
8. Murray C, Tourtellotte A, Lipson D, Wysocki A. Medicaid long-term services and supports annual expenditures report: federal fiscal years 2017 and 2018. Mathematica. January 7, 2021. Accessed November 11, 2022. https://www.medicaid.gov/medicaid/long-term-services-supports/downloads/ltssexpenditures-2017-2018.pdf
9. Dobson C, Gibbs S, Mosey A, Smith L. Demonstrating the value of Medicaid MLTSS programs. Center for Healthcare Strategies. May 12, 2017. Accessed November 11, 2022. https://www.chcs.org/media/FINAL-Demonstrating-the-Value-of-MLTSS-5-12-17.pdf
10. Atherly A, Florence C, Thorpe KE. Health plan switching among members of the Federal Employees Health Benefits Program. Inquiry. 2005;42(3):255-265. doi:10.5034/inquiryjrnl_42.3.255
11. Jacobson G, Neuman T, Damico A. Medicare Advantage plan switching: exception or norm? Kaiser Family Foundation. September 2016. Accessed April 25, 2022. https://files.kff.org/attachment/Issue-Brief-Medicare-
12. Schlesinger M, Druss B, Thomas T. No exit? the effect of health status on dissatisfaction and disenrollment from health plans. Health Serv Res. 1999;34(2):547-576.
13. Department of Medical Assistance Services. Virginia.gov. Accessed September 9, 2021. https://www.dmas.virginia.gov/
14. Commonwealth Coordinated Care Plus. Virginia Commonwealth University Department of Health Behavior and Policy. Accessed September 9, 2021. https://hbp.vcu.edu/policy-briefs/commonwealth-coordinated-care-plus/
15. Rach J, Baxter M, Quave D, Yount N, Shaller D. CAHPS health plan survey database 2019 chartbook. Agency for Healthcare Research and Quality. October 2019. Accessed April 10, 2022. https://cahpsdatabase.ahrq.gov/files/2019CAHPSHealthPlanChartbook.pdf
16. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev. 1977;84(2):191-215. doi:10.1037//0033-295x.84.2.191
17. Gruber-Baldini AL, Velozo C, Romero S, Shulman LM. Validation of the PROMIS® measures of self-efficacy for managing chronic conditions. Qual Life Res. 2017;26(7):1915-1924. doi:10.1007/s11136-017-1527-3
18. Kan K, Barnes AJ, Hanoch Y, Federman AD. Self-efficacy in insurance decision making among older adults. Am J Manag Care. 2015;21(4):e247-e254.
19. Garnick DW, Hendricks AM, Thorpe KE, Newhouse JP, Donelan K, Blendon RJ. How well do Americans understand their health coverage? Health Aff (Millwood). 1993;12(3):204-212. doi:10.1377/hlthaff.12.3.204
20. Rossiter LF, Langwell K, Wan TTH, Rivnyak M. Patient satisfaction among elderly enrollees and disenrollees in Medicare health maintenance organizations: results from the National Medicare Competition Evaluation. JAMA. 1989;262(1):57-63. doi:10.1001/jama.1989.03430010069033
21. Wray CM, Khare M, Keyhani S. Access to care, cost of care, and satisfaction with care among adults with private and public health insurance in the US. JAMA Netw Open. 2021;4(6):e2110275. doi:10.1001/jamanetworkopen.2021.10275
22. Tavares AI, Ferreira PL. Public satisfaction with health system coverage, empirical evidence from SHARE data. Int J Health Econ Manag. 2020;20(3):229-249. doi:10.1007/s10754-020-09279-x